Currently, conventional coaxial cable TV systems can deliver tens of megabits per second data transmissions downstream (i.e., from the headend unit to various subscribers). Also presently feasible is the use of diplexing amplifiers and reverse amplifiers to transmit information upstream (i.e., from one or more subscribers back to the headend unit) on the common coaxial cable as well. However, to activate this reverse channel entails a major fixed cost of adding amplifiers and filters and balancing the upstream gain. If there are many customers for the new two-way services, then the cable operator's investment in activating the two way capability can be justified. However, if there are but relatively few scattered upstream services customers, then the cable operator faces a significant fixed cost to activate the two way capability that cannot be fully amortized over the few customers. This is a "chicken and egg" type problem. On the one hand, the desired solution requires access before the market can be proven, but the access comes at the expense of an often unacceptable front end cost.
In an effort to mitigate cost, various systems have been proposed in which distinct, separate paths were established for downstream versus upstream communications. The use of a separate telephone line return for the upstream cable subscriber signals is well known and is used in a number of different cable modems. The idea of heterodyned frequency translation per se is also old art. And the use of frequency translation in cable systems is probably best developed in the RAD concept used to transmit PCS signals over a cable system. The RAD concept includes the use of inexpensive antennas and frequency shifters as a two way delivery mechanism to remote PCS devices. Over-the-air, two-way signals from cordless Personal Communications Service telephones, say at 1800 MHz from a number of remote antenna sites are frequency shifted and carried to and from one or more central sites.
Although these alternative prior art systems offer some degree of flexibility and enhanced functionalities, they nonetheless suffer from several disadvantages. Today, only 10-15% of the cable plants have been converted to two-way operation. Where telephone lines are used for upstream transmission, the upstream paths are seriously bandwidth limited. And with such approaches, a different protocol is needed in the upstream which requires implementation of different additional hardware at both the cable TV headend and at the subscriber's premises. Scalability and upgradeability were also potential problems with some prior art systems. In addition, there might be serious compatibility issues to be overcome.
Alternative prior art in wireless telephony does include treatment of multiple receiver antennas but used only in a two-way wireless telephony where synchronous transmission is required over the wireless links such that time synchronization and signal formats are compatible with traditional telephony time-slot interchange trunk systems. These systems do not predict the use of a one-way transparent wireless paths in cable television upstream distribution systems for the support of asynchronous packet transmission, nor do they predict the transparent behavior of this invention in that cable modems located on both wired and wireless paths may be simultaneously received on the same upstream channel by a headend cable modem controller.
Alternative prior art in asymmetric cable modem systems universally have restricted the upstream wireless link to a single cable modem channel. These systems suffer both from scaling issues and from restricting deployment in geographic locations to situations where RF signal propagation is only favorable to a few solutions thereby restricting broad deployment needed for a successful cable modem business. These systems do not predict the need for multiple receiver antenna sites and subsequent selection and packet duplication avoidance issues required to meet low cost, nor do they predict the transparent behavior of this invention in that cable modems located on both wired and wireless paths may be simultaneously received on the same upstream channel by a headend cable modem controller, nor do they predict the need for simultaneous carriage of multiple cable modem upstream channels needed to scale to deployment size needed for viable business.
In light of these limitations of the prior art, there exists a need in the cable industry for an efficient, transparent, and cost-effective approach for converting standard one-way cable systems into two-way systems. The present invention provides a solution whereby a transparent radio return path is used for providing upstream communications, which allows cable operators to be able to selectively upgrade their systems on an as-needed basis. Costs are minimized by leveraging the same frequency, scheduling, and modulation schemes to keep the conversion from the radio path signal to the signal on the cable as simple and straightforward as possible.